I noticed that every time I submit my code for a multiplayer level, there is an immediate drop on my scores. For example, more than 500 points drop here, slowly building up again, even with 100% wins.
The other issue that bothers me is the lack of games. For example: code submitted more than a day ago, only one game played since then:
Is there a reason for these “anomalies”?
I can’t say what is happening in the first picture, but the second picture shows a known flaw in the matchmaking system. If you submit code and someone else immediately after that submits code, the probability that your game is chosen for a match is tiny.
I believe the devs are already working on it.
J_F_B_M covered the second question.
The first is because you submitted new code . . . Think of the new code is it a completely new person or just a small improvement? . . . Is it a small change, is it a big change, did you blow the old code away and submit all new code, did you make a typo??? The system doesn’t know if it should trust your new code or not so you have to earn back part of your previous reputation (being that the system doesn’t know what you did to your code it would be just as fair to count each submission as if it was the first. So, in that sense it is especially nice of “them” to only give a small knock to your OLD score.)
So your score is not calculated based on your win/lose ratio? Could someone explain how it’s calculated then? Or is it already described somewhere?
I thought that - for example - winning 95 games out of a 100 gets you a score of e.g. 5000. So if your new code is similarly good - let’s say 98 wins out of 100 - your score continues to improve from there… Apparently, it’s not the case.
There are several posts that touch on this subject if you were to search (I used “rank”). This exact question has been asked (and answered) before, see this post.
Thanks for the answer, looks like I need to improve my searching/reading skills 
Can I close the topic somehow?
This topic is now closed. New replies are no longer allowed.